lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <CAADnVQLtjnp+Rk5Zof5WHQ2Sw-xYxY39OqYgr_DKDgS8M+5HEg@mail.gmail.com>
Date:   Thu, 21 May 2020 17:49:24 -0700
From:   Alexei Starovoitov <alexei.starovoitov@...il.com>
To:     Andrii Nakryiko <andrii.nakryiko@...il.com>
Cc:     John Fastabend <john.fastabend@...il.com>,
        Yonghong Song <yhs@...com>,
        Alexei Starovoitov <ast@...nel.org>,
        Daniel Borkmann <daniel@...earbox.net>,
        Networking <netdev@...r.kernel.org>, bpf <bpf@...r.kernel.org>
Subject: Re: [bpf-next PATCH v2 0/4] ] verifier, improve ptr is_branch_taken logic

On Thu, May 21, 2020 at 3:21 PM Andrii Nakryiko
<andrii.nakryiko@...il.com> wrote:
>
> On Thu, May 21, 2020 at 1:07 PM John Fastabend <john.fastabend@...il.com> wrote:
> >
> > This series adds logic to the verifier to handle the case where a
> > pointer is known to be non-null but then the verifier encountesr a
> > instruction, such as 'if ptr == 0 goto X' or 'if ptr != 0 goto X',
> > where the pointer is compared against 0. Because the verifier tracks
> > if a pointer may be null in many cases we can improve the branch
> > tracking by following the case known to be true.
> >
> > The first patch adds the verifier logic and patches 2-4 add the
> > test cases.
> >
> > v1->v2: fix verifier logic to return -1 indicating both paths must
> > still be walked if value is not zero. Move mark_precision skip for
> > this case into caller of mark_precision to ensure mark_precision can
> > still catch other misuses. And add PTR_TYPE_BTF_ID to our list of
> > supported types. Finally add one more test to catch the value not
> > equal zero case. Thanks to Andrii for original review.
> >
> > Also fixed up commit messages hopefully its better now.
> >
>
> Yeah, much better, thanks! Few typos don't count ;)
>
> For the series:
>
> Acked-by: Andrii Nakryiko <andriin@...com>

Applied. Thanks a lot everyone.

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ